Download Advances in Natural and Applied Sciences

Advances in Natural and Applied Sciences, 8(17) Special 2014, Pages: 23-35
AENSI Journals
Advances in Natural and Applied Sciences
ISSN:1995-0772 EISSN: 1998-1090
Journal home page: www.aensiweb.com/ANAS
A Survey on Cross-Layer Architecture with QoS Assurances for Wireless Multimedia
Sensor Networks
1
A. Sivagami, 2S. Malarkkan, 3M. Asvini Devi
Research Scholar, Sri Chandrasekharendra Saraswathi Viswa Mahavidyalaya University, Kancheepuram Dt. – 631 561, India and
Assistant Professor in Department of Electronics & Communication Engineering, Sri Manakula Vinayagar Engineering ollege, Puducherry605107, India
2
Professor & Principal, Manakula Vinayagar Institute of Technology, Puducherry – 605 107, India
3
PG Scholar, Department of Electronics & Communication Engineering, Sri Manakula Vinayagar Engineering College, Puducherry605107, India
1
ARTICLE INFO
Article history:
Received 3 September 2014
Received in revised form 30 October
2014
Accepted 4 November 2014
Keywords:
Cross- Layer design, WMSN,
Optimization of Protocols
ABSTRACT
Growing interest and diffusion of wireless networking technologies is underlining new
challenges in the design and optimization of communication protocols. Usually,
protocol architectures follow strict layering principles, which make sure
interoperability, fast deployment, and efficient implementations. On the other hand,
need of organization among layer restrictions the performance of such architectures due
to the specific challenge posed by wireless nature of the transmission links. To
overcome such limitations, cross-layer design has been proposed. Its core idea is to
maintain the functionalities associated to the original layers but to allow coordination,
contact and joint optimization of protocols crossing different layers. This chapter
introduces the reader by means of the notion of the cross-layer design, exactness
motivations and requirements, presents the main building blocks enabling collaboration
between layers, and compares available signal architectures. Then, after mention
current position of equality activities in the field, it presents novel architectural
solutions connecting cross-layer design which are proposed in the framework of after
that making network infrastructure.
© 2014 AENSI Publisher All rights reserved.
To Cite This Article: A. Sivagami, S. Malarkkan, M. Asvini Devi, A Survey on Cross-Layer Architecture with QoS Assurances for
Wireless Multimedia Sensor Networks. Adv. in Nat. Appl. Sci., 8(17): 23-35, 2014
INTRODUCTION
Wireless Sensor Networks (WSN) has recently been the meeting point of a major quantity of thought and
attempt of the investigate centre of population (D. G. Costa, 2011). The main driving force has been tackle to
face the fake by WSN paradigm, i.e., partial node power, dispensation, and statement ability, dense set-up
operation, multi-hop communications, and mixed application-specific requirements. The huge majority of this
revision applies to conservative WSN request which needs dependable and skilled message of scalar event kind
and sensor information such as heat, force, moisture. With the ease of access of low-cost small-scale imaging
sensors, CMOS cameras, microphones, which may all in excess of the put capture CD content from the field,
Wireless Multimedia Sensor Networks (WMSN) have been intended and drawn the straight thought of the study
area. WMSN submission, e.g., compact disc surveillance set-up, target way, green monitor, and traffic
organization systems, require effectual harvesting and message of event features in the form of CD such as
audio, image, and video. To this end, additional test for energy-efficient compact disc handing out and statement
in WMSN (H.-P. Shiang, 2009). The estimate and statement infrastructure related with sensor networks is often
exact to this situation and rooted in the device and application-based nature of these networks. For case, unlike
most other settings, in-network giving out is popular in sensor networks; still, node power (and/or battery life) is
a key aim thought. The information collected is typically parametric in nature, but with the coming out of lowbit-rate video [e.g., Moving Pictures Expert Group 4 (MPEG-4)] and imaging algorithms, some systems also
hold up these types of media. We provide a show of the basic aspects of wireless sensor networks (WSNs) (J. B.
Othman, 2010).
With the recent advance in micro electro mechanical systems (MEMS) technology(N. A. Ali,2008),
wireless communications, and digital electronics, the design and increase of low-cost, low-power,
multifunctional antenna nodes that are small in size and correspond untethered in small detachment have
Corresponding Author: A. Sivagami, Assistant Professor in Department of Electronics & Communication Engineering,
Sri Manakula Vinayagar Engineering College, Puducherry- 605 107, India
E-mail: [email protected]
24
A. Sivagami et al,2014
Advances in Natural and Applied Sciences, 8(17) Special 2014, Pages: 23-35
become feasible. Recently, considerable amounts of research efforts have enabled the actual execution and use
of sensor networks tailored to the unique requirements of certain sensing and monitoring applications. In order
to realize the existing and potential applications for WSNs, sophisticated and extremely efficient communication
protocols are required. WSNs are collected of a large number of antenna nodes, which are densely deployed
either inside a physical occurrence or very close to it. The intrinsic properties of individual sensor nodes pose
additional challenges to the statement protocols in terms of energy consumption. WSN applications and
communication protocols are mainly tailored to provide high energy efficiency. Sensor nodes carry limited
power sources. Therefore, while traditional networks are designed to improve performance metrics such as
throughput and delay, WSN protocol center primarily on power protection. The use of WSNs is another factor
that is considered in developing WSN protocols (Shu, L., 2010). The place of the sensor nodes need not be
engineered or prearranged. This tolerates random deployment in inaccessible terrains or disaster let go process.
On the additional hand over, this random operation requires the development of self-organizing protocol for the
communication protocol stack. In addition to the placement of nodes, the breadth in the network is also
exploited in WSN protocols (I. T. Almalkawi,2010). As a result, the spatial-temporal correlation-based process
emerges for better ability in system wireless sensors.
Cross layer design (cld) definition:
To fully optimize wireless broadband networks, both the dispute from the objective medium and the QoSdemands have to be taken into explanation (M. Sharif, 2005). Rate, power and code at the material layer can be
modified to meet the provisions of the request given the accessible guide and net state. Data has to be common
between (all) coating to obtain the utmost likely adaptively (Lecuire, V, 2008)
Cross-Layer Design:
The enclosed method stack mean is highly rigid and firm and each sheet only takes care about the layer
openly above it or the one openly below it. This fallout in non-collaboration which exists between dissimilar
layers, in fact since no one at that time saw any need for such a non-collaborative sketch known as the crosslayer design (Physical Layer Standard,1999). The CLD move in the path of to net stack plan is in the past a big
shift in how one designs a message system. Not only do the request, protocols and hardware need to be
reemployment to be able to support the new addition, but the entire design of CLD confront all engineers and
researchers know about system protocol, coating (A. C. Begen,2005) stack plan and system building. While
CLD might seem basic instead of evolutionary at first, there are implementations (Licrandro, 2008) accessible
today which tries to comprise some of the key basics of the CLD approach into on hand procedure and film.
Fig.1: Example reference architecture with Defined Interface.
The Evolutionary Approach to CLD:
An evolutionary approach to CLD (Siteplanner) always seeks to make bigger the available enclosed
structure, in order to keep computability. Most CLDs today are evolutionary (G. Yu, 2007), because talent with
existing systems and networks is very significant both for end users and viable actor. The reason for this is quite
simple. Since an evolutionary CLD is always bound by its unique strict covered structure, an addition of this
will also always be faulty. Most often only two or maybe three layers need to share in classify (M. A. Yigitel,
2011), and thus it’s also rational to extend the creative strict covered makeup.
25
A. Sivagami et al,2014
Advances in Natural and Applied Sciences, 8(17) Special 2014, Pages: 23-35
The Revolutionary Approach to CLD:
A radical approach to CLD, or any plan for that matter, is not jump by an obtainable implementation, and as
such does not need to cooperation to maintain ability (E. Hossain, 2004). Since most CLDs today are
evolutionary, it’s hard to and case of a revolutionary CLD. We take a look at Shannon mappings, which also is
CLD in a radical since (Daeyoung Park, 2005). These penalty and ideas might give us a new kind of a exact
difficulty and may be practical to existing evolutionary implementations as well the past has exposed many
times, that care an open mind about new and ground-breaking thoughts is evermore major (T. He, 2003).
ISO/OSI TCP/IP Protocol Stack Principles:
Where an evolutionary CLD approach prioritizes capability rest and act later, an original design does the
different (W. Kumwilaisak, 2003). A revolutionary approach can however be applied to highly specific
problems where backwards potential is not important. For example, we take a look at the Wireless Sensor
Network (WSN) setting, where this approach can be applied (Shu, 2010). At present, design of network
architectures is based on the layering principles(S. Kunniyur,2000), which provide a smart tool for scheming
interoperable systems for fast operation and efficient implementation.ISO/OSI model was urban to support
promptness of network architectures using the layered model. The main concept inspiring layering is the next:
• Each layer performs a subset of the necessary letter function.
• Each film relies on the next subordinate layer to affect more ancient functions (Shu, 2010).
• Each layer supply services to the after that higher layer.
• Change in one layer is supposed not to require change in other layers.
Such concept were used to define a position protocol stack of seven layers (H. Kimand, 2009), going from
the physical layer (concerned with spread of a formless stream of bits over a message channel) up to the request
layer (providing access to the OSI environment). A procedure at a given layer is implementing by a (software,
firmware, or hardware) entity, which communicate with other entities (on other networked systems)( M. Chen,
2008) implement the same protocol by Protocol Data Units (PDUs). A PDU is built by goods (data addressed or
generated by an entity at a higher nearby layer) and header (which contains protocol information). PDU
arrangement as well as service meaning is exacting by the protocol at a given level of the stack. The same
concept is at the basis of the de-facto normal protocol stack on the Internet, namely the TCP/IP protocol stack.
The main gain increase from the layering example is the modularity in protocol design, which enables
interoperability and improved design of message protocols (X. Lin, 2005). Moreover, a protocol within a given
layer is explain in terms of functionalities it offers, while execution details and interior parameter are hidden to
the rest layer (the so-called “in order hiding” property) (F. Hou, 2006).
Encoding standards:
A PHY-MAC Cross-Layer Optimization for Wireless Broadband Network:
Triantafyllopoulou, Passas, and Kaloxylos have potential a cross-layer optimization device for multimedia
traffic over IEEE 802.16 system. The system utilize in order provided by the PHY and MAC layers, such as
signal value, package loss speed and the suggest holdup, in order to manage limit at the PHY and request layers
and get better the act of the system(V. Srivastava,2005). Mostly, the adaptive modulation ability of the PHY
layer and the multi-rate data-encoding characteristic of solid disk request are joint to achieve a better end-user
QoS (Physical Layer Aspects,2001). The cross-layer optimizer is split into two parts- the BS part and the SS
part, residing at the base station and the subscriber stations respectively (Q. Xu, 2007). The BS part accepts an
abstraction of layer-specific information about the channel conditions and QoS parameter of active connections,
provided by the BS PHY and MAC layers. Based on this information, a specific decision algorithm determines
the most suitable modulation and/or traffic rate of each SS, separately for each direction (uplink and downlink)
(J. Villal, 2007). Finally, the BS part informs the matching layers of the required modifications. If the decision
of the BS part involves traffic rate changes, it converse with the SS part through SS MAC layer, which instructs
the SS application layer accordingly. In the proposed architecture, the SS part is planned as a passive unit that
only instructs the SS application layer based on BS part’s idea.
Adaptations at Different Layers of the Protocol Stack:
Triantafyllopoulou, Passas, and Kaloxylos have future a cross-layer optimization device for CD traffic over
IEEE 802.16 system (V. T. Raisinghani, 2006). Fundamentally, the adaptive modulation capability of the PHY
layer and the multi-rate data-encoding characteristic of compact disk application are joint to achieve an
improved end-user QoS(I. F. Akyildiz,2007). The cross-layer optimizer is split into two parts- the BS part and
the SS part, residing at the base station and the subscriber stations respectively (D-K. Triantafyllopoulou, 2007).
The BS part accepts an abstraction of layer-specific information about the channel conditions and QoS
parameters of active connections, provided by the BS PHY and MAC layers (F. Oldewurtel,2008). Based on this
information, a specific decision algorithm determines the most suitable modulation and/or traffic rate of each
SS, separately for each direction (uplink and downlink) (S. Kunniyur, 2000). Different types of adaptation are
26
A. Sivagami et al,2014
Advances in Natural and Applied Sciences, 8(17) Special 2014, Pages: 23-35
required at different layers of the normal protocol stack for providing a robust QoS support to multimedia
application over wireless networks(Q. Wang,2003), it has by now been seen that wireless channel pose a
number of challenges in this aspect.. It is known that RTP/UDP/IP and TCP/IP have the problem of large header
overhead on bandwidth-limited links (S. Shakkottai, 2003). Error protection, power saving, and stand-in
management are some of the important issues to be handled in the application layers. These layer specific issues
are described in details in the next subsections (IEEE Standard 802.16 Working Group, 2002).
Link Layer Adaptation Mechanisms:
There are a number of currently existing approaches for link layer adaptation under varying wireless
channel conditions (D. L. Goeckel, 1999). The important ones in this group are: (i) application adaptive ARQ,
(ii) priority-based scheduling, (iii) channel aware scheduling. These are described in detail in the following.
Application Adaptive ARQ: To beat packet loss, a method called Automatic Repeat Request (ARQ) is used for
packet retransmissions (A. Boulis, 2003). ARQ uses acknowledgments and timeouts to achieve reliable data
transmission. The handset sends an acknowledgement (ACK) to the teller to indicate that it has correctly
received a data edge or packet (George Xylomenos,2001). The sender waits for a predefined period (timeout)
for the ACK to arrive. If ACK arrives then the sender sends the next packet. Or else, it resends the earlier packet
until it receives an ACK or exceeds a predefined number of retransmissions (Q. Liu, 2004). ARQ can be
implementing at the application/transport layer as well as the link-layer. Link-layer ARQ is more effective than
application/transport layer ARQ because -(I) it has a shorter control loop and hence can pick up lost data more
quickly, (ii) it operates on frame that are much lesser than the IP datagram’s and (iii) it This in order can include
in order about the state of the link and waterway(Z. Xiong,2004). However, best performance cannot be
achieved using link-level ARQ as it may result in an undesirably large amount of data retransmission among
different layers and consequent presentation deprivation in transport protocols. A more effectual way of using
the link-layer ARQ is to make it aware of the application QoS on a per packet basis. The link-layer ARQ can
then adjust its performance accordingly.
a) Priority-based Scheduling: In priority-based schedulers, packets are group into several classes with dissimilar
priority according to their QoS supplies i.e. the MAC layer is made aware of the application layer QoS. Packets
belong to higher priority classes are more likely to be transmitting first. Packets in the same class are served in a
FIFO manner. Based upon the main concern development device, each QoS class will have some sort of
arithmetical QoS guarantee.
b) Channel-aware Scheduling: In a manifold contact wireless network, the radio direct is normally distinguish
by time-varying fading. To exploit the time-varying point, a type of channel-state needy growth, called
multiuser diversity, can be not working to recover system act. For a wireless letter system with numerous MSs
having free time-varying vanishing direct, we can presume that the guide are either ON i.e. one packet can be
transmitted fruitfully to the mobile user during the time-slot or OFF i.e. unsuitable for transmission. The
scheduler at the BS MAC layer gets the channel state information from its PHY layer. The scheduler at the BS
transmits to a user whose channel is in the ON state. In case more than one user channel is in ON state, the
scheduler selects a user channel randomly. No data is sent by the BS when the entire channel is OFF. For a 3user case, all the channels will be in OFF state only for 1/8 of the time on average. Thus, total data rate achieved
by the scheduler is (1-1/8) = 7/8 packets per slot. Hence typical data rate per user is (7/8)/3 = 7/24 packets/slot.
For round-robin scheduling with 3 users, each user will get 1/3 slot time. Since the user channels are equally
likely to be ON or OFF in each timeslot, each user will get a data rate of (1/3)/2 = 1/6 packets/slot which is
about half that of the channel-aware multi-user diversity scheduler. Thus, overall resource operation can be
better by using channel-aware scheduling mechanism (S. Pollin, 2003). Then a distortion-minimized bit share
scheme with hybrid uneven error safety (UEP) and delay-constrained automatic repeat request (ARQ) that with
passion adapts to the estimated time-varying set of connections conditions is future. The future scheme can
much improve the reconstructed video quality under ruined system setting.
Transport Layer Adaptation Mechanisms:
Due to the channel variations, packet losses are predictable at the receiver of the wireless message system
In order to deliver multimedia over wireless networks, it’s necessary to estimate the condition of the underlying
network so that the strict QoS constraint for multimedia applications can be adhered to. Blockage may occur
within a network when routers are overloaded with traffic which in turn causes their queues to build up and
eventually overflow, leading to high delays and packet losses (T. Rappaport, 2002). Network conditions can be
assessed mainly through congestion estimation based on - packet loss and current available bandwidth; TCP
interprets all losses within a network as being congestion related. This is mainly because TCP was originally
designed for wired networks with a reliable physical layer, where packet loss mainly results from set-up
blockage. This quality of TCP is unsuitable for wireless networks since losses due to natural channel errors are
also treated as a signal of network blocking. This causes the source TCP to reduce its transmission rate by
decrease its congestion window size, even though there is no network congestion resulting in avoidable cut in
27
A. Sivagami et al,2014
Advances in Natural and Applied Sciences, 8(17) Special 2014, Pages: 23-35
throughput. In principle, packet loss due to channel errors should result in retransmissions not rate reduction. In
order to improve the TCP performance in wireless scenario, it is necessary to differentiate the congestion-related
packet losses from non-congestion packet losses. Two methods to achieve this objective - watch TCP and TCP
with ECN are described below. Snoop TCP: Snoop TCP gives a dependable TCP-aware link layer TCP with
open Congestion Notification (ECN): ECN is an end-to-end device to notify the sender whenever obstruction
occurs in the network. In a typical IP packet header, an ECN field is included. When a router detects persistent
(not transient) congestion in the complex, it sets the ECN field and the packet is said to be marked. The marked
packet eventually reaches the destination, which in turn informs the source about the congestion by setting the
ECN-Echo flag in the TCP heading format. In the intended protocol, named Image Transport Protocol (ITP),
application data unit (ADU) limits are exposed to the transport module. This enables the convey module to
perform out-of-order release of packets. As the move is aware of ask for surround limits, the projected advance
expand on the application-level framing (ALF) attitude, which offer a one-to-one charting from an ADU to a
network small package or procedure data unit (PDU) ( J. G. Kim,2000). ITP deviate from the TCP’s notion of
reliable delivery. Instead, it incorporates selective loyalty, where the receiver is in control of deciding on what is
to be transmitted from the sender at any instant. ITP runs over UDP, incorporate receiver-driven selective
reliability, and uses a overcrowding manager (CM) to become accustomed to set-up congestion. It also enable a
variety of new receiver post-processing algorithms such as error cover up that extra get better the interactivity
and receptiveness of recreate imagery.
Application Layer Adaptation Mechanisms:
Due to concurrent nature, multimedia services classically require QoS guarantees like large bandwidth,
strict delay bound and relatively error-free video/audio/speech quality. Multimedia services over the wireless
channels become very testing due to the dynamic uncertain nature of the channel resultant in variable available
bandwidths and casual packet losses. The main objectives of the application layer QoS control for compact disk
communication over wireless network are ,(i) to avoid burst losses and excessive delay (caused by network
congestion) that have a upsetting effect on multimedia presentation quality, and (ii) to make the most of
multimedia quality even when packet loss occurs in a wireless communication network. In traditional layered
design approach, source rate control and congestion control are designed independently and in isolation with
each other (P. Zhu, 2007). This imposes a control on the overall system performance e.g., end-to-end delay
constraint and smooth playback quality. Obstruction control for streaming multimedia usually needs to smooth
its sending rate to help the application achieve flat playback quality. However, this is not always possible as the
source code block at request layer can change the coding complexity and sending rate suddenly based on its
QoS requirements, unless notified by the transport layer. Moreover, source rate control alone cannot agreement
the end-to-end delay limitation due to minimum bandwidth requirement and quality softness requirement. The
architecture of the system is depicted in. The mechanism works as follows. If the transfer rate is made to
temporarily violate the TCP-friendliness nature of the transport layer, the quality of the compact disk content is
significantly improved. The long-term TCP– friendly sending rate is preserved by implement the rate return
algorithm. At the request layer, the near system buffer running device is used to translate the QoS supplies of the
request in terms of the basis and distribution rates. There is a middleware part situated sandwiched between the
request layer and the carry layer wherein the dual conclusion of the source rate and the sending rate is done. To
make their planned mechanism work effectively in wireless environment, the authors have included the logical
Rate Control (ARC) protocol(B. Henty,2001) presented in ARC protocol is planned to achieve high throughput
and compact disk support for real-time traffic flows while preserve fairness to the TCP sources sharing the same
wired link resources. The sender performs rate control using the ARC equation to avoid any unnecessary rate
reduction due to wireless link errors, thus enabling the system to work properly in a wireless environment. Joint
aim of Source Coding and Link Layer FEC/Retransmission: In order to adapt to the varying network conditions
like loss, delay, variable bandwidth etc., the media codes are designed using scalable coding Techniques.
Scalability in video can be achieved by layered coding technique (e.g. in MPEG-4). The covered coding
technology divide the video into several layers and the incremental reception of layers increases the media
fidelity. Video codec’s encode a video sequence into one base layer and multiple enhancement layers based on
any of the following three classes of layered coding techniques – temporal, spatial and signal-to-noise ratio
scalability. Forward error correction (FEC) and retransmissions are major link layer error correction
mechanisms. FEC is a channel coding technique protecting the source data by adding disused data during
transmission. Thus FEC is not bandwidth well-organized but very effective in application with strict delay
requirements such as voice infrastructure (retransmission may induce huge latency). Applications where delay
requirements are much relaxed, link layer retransmissions are more suitable as it is bandwidth efficient unlike
FEC. As mentioned earlier, in a multi-hop wireless scenario, packet losses can occur due to network congestion
or wireless transmission errors, which invariably will have different loss patterns. According to, such different
loss patters will get reflected as different perceived QoS at the application layer (S. Pollin, 2003). A loss
differentiated rate-distortion based bit part scheme is proposed in that minimizes end-to-end video distortion
28
A. Sivagami et al,2014
Advances in Natural and Applied Sciences, 8(17) Special 2014, Pages: 23-35
taking the different loss patterns into account, which shows that both basis coding and channel coding
parameters can affect the final media quality(L. Chen,2005). Joint Source and Channel Coding (JSCC) schemes
are proposed to achieve the optimal end-to-end quality by adjusting the source and direct coding parameters
simultaneously. A simple JSCC scheme using UEP is presented by Jiang, Zhuang and Shen. UEP can be
performed with Bose-Chaudhuri-Hocquenghem (BCH) codes, Reed-Solomon (RS) codes, and Rate-Compatible
Punctured Convolution (RCPC) codes with different coding rates for packets with different priorities.
Multimedia-based cross-layer optimization in vsns:
Recent chart sensor network are strongly incomplete by the take on coding technique. The source plan rate,
the power use over the existing path(s) and the error-resilience of the statement depend on the way still images
and string is set and decode. The organization of the multimedia-based cross-layer optimization solutions in
visual feeler network in two categories:
a)Image-based: Images are snapshots rescue by video-based sensors, with respect to the field-of-view of the
preset camera at each sensor. The image monitor request can be delay free or involve real-time transmission,
direct impacting the approve report building. A dissimilar key aspect is that images typically require less
bandwidth than video media, also overriding less energy due to lower plan rates (G. Yu, 2007).
b) Video-based: Some request honestly advantage from video media to provide an improved kind of the
monitored end or scene. The same limit in terms of delay and quality considered for image-based solutions are
valid for video, with adding up of some very pertinent independence, such as the number of transmit frames per
second. The character of video media imposes more strict requirements for delay and jitter, also difficult more
bandwidth than imagery. In such background, design issue as multipath steering get even more significance.
Besides the type of average, we can classify the cross-layer optimization in visual sensor networks
according to where the coding is performed:
c)Source processing: When raw chart data are fully process by the sensor which together them, the density is
said to be perform locally at the source node. For that, the source must have enough handing out and memory
property to right carry out the training algorithms. Source processing potentially reduces the end-to-end delay,
but may exact large force use from the source node.
d)In-network processing: When basis nodes do not contain sufficient resources to program a large quantity of
data, when the note structure define specialized tasks for center nodes or if it is preferred to save force of basis
nodes, in-network giving out is performed. It must be precise some protocol to share the original data in the
middle of the nodes which will apply pressure the facts (A. C. Begen, 2005). In such case, the vigor is saved in
basis nodes at the cost of extra complexity. The next subsections survey multimedia-based cross-layer
optimization allowing for all the earlier on hand over aim issue.
Image-Based Cross-Layer Optimization:
When commerce with picture density, three very pertinent self will be typically present in the encoded
images: unequal meaning, error tolerance and forced error spread (S. Ehsan, 2011). Analyze the surveyed
works; we can build out unlike approach for cross-layer optimization. Among the projected solutions, waveletbased prioritization and progressive coding seem to be very fit for wireless image sensor networks, and new
investigations regarding cross-layer design employ such code practice should still arise. The image-based crosslayer optimization likely in is a priority driven scheduling algorithm. That algorithm schedules for program
more data from key sub-images, according to the accessible bandwidth. Additionally, when it is impossible to
transmit the entire picture within the three bandwidth and power constraint, some data pertaining to insignificant
sub-images are surplus. VSN demand can profit for visual monitor where priority-based JPEG training like the
one existing in is used in basis nodes. For example, if application are monitor an affecting object, such a coding
strategy would potentially reduce the bandwidth rations (and energy consumption) of transmission of lessrelevant parts of the scene (as visual data of the sky and the ground), keeping the viewing of the target with high
quality for the application. Besides multipath routing, the authors propose the use of Unequal Error defense to
offer an added level of stability.
Video-Based Cross-Layer Optimization:
Video program on visual sensor network is even tougher than the image-based program. In fact, video
requires more bandwidth than image does, and the amount of data to be coded is much larger when source nodes
are stream video (J. B. Othman, 2010). The coding process of collected video may follow different approach,
achieving high or low compression. In video-based cross-layer optimization, protocols and algorithms in
transport, network, MAC and even physical layers may exploit in order of the coded data (application layer) to
reduce the transmission rate, to decline the end-to-end delay, to protect more relevant data against bit errors and
packet dropping and to save energy in source nodes and throughout the network(L. Shu,2009). There are two
major contributions in. In first place, the AC2 queue may overflow if many video packets are being received,
resulting in packet dropping. Such blocking is addressed by a choosy algorithm that drops packets based on the
29
A. Sivagami et al,2014
Advances in Natural and Applied Sciences, 8(17) Special 2014, Pages: 23-35
relevance of the fixed data according to the MPEG-4 coding technique, reducing the impact on the perceptual
quality of the received video(R. Zhang,2000). The second proposed solution tries to avoid congestion by
performing queue management in go forward. If the current size of AC2 queue is higher than a entrance,
incoming (forward) packets are allocated to AC1 queue, while locally shaped packets are assigned to AC0
queue. Such system assures that forward packets have higher transmitted priority than local packets. The work
in proposes a multimedia-based cross-layer regarding the IEEE 802.11s MAC protocol. In fact, this MAC
protocol is mainly anticipated for mesh networks, where energy is not a major fear as in WMSNs. Although
authors indicate the wished-for solution for wireless multimedia sensor network, supplementary investigation
should be performed to assess the chance of IEEE 802.11s in energy-constrained low-cost wireless sensor
networks. Other works in the literature investigate multimedia-based cross-layer optimization focusing MAC
protocol IEEE 802.11e, as for H.264 and for MPEG-4. However, neither of these works considers energy as a
key design issue. Algorithms to prioritize packet according to their implication for the translate process are very
talented for VSN application, achieving relevant results in end-to-end delay, perceived visual quality and energy
consumption(S. Kompella,2007). In recent years, such techniques have been exploited for multipath routing,
especially when sensor nodes collect visual and audio data from the monitor field. The context-aware
multimedia-based cross layer optimization scheme proposed in exploits multipath steering along with the
relevance of the fixed data for able path selection regarding the end-to-end message delay. It is not properly
specified a coding method, but the future routing scheme is very suitable for multiple account coding. To
facilitate effort define the MPMPS (Multi-Priority Multi-Path Selection) algorithm to find the paths with lower
end-to-end delay for multimedia stream in WMSNs, way in mind a set of available node-disjoint paths. Such
paths are node-disjoint when they have no ordinary middle nodes. The node-disjoint paths are bare employ the
Two-phase Geographic insatiable forward (TPGF) algorithm, which finds the most number of optimal node
disjoint routing paths in terms of path length and the end-to-end transmission delay, potentially benefit wireless
CD sensor networks request. The authors argue that conventional multipath direction-finding protocols do not
provide a potent probing apparatus to find the multiple optimized direction-finding paths and to bypass holes.
Fig. 2: Multipath multimedia transmission through three-nodes disjoint path
The work easy to get to in propose the split of the source stream into image and audio sub streams, giving to
each resulting sub stream a particular main concern according to the current monitoring being perform. The path
with inferior delay is assigned to the higher main concern sub streams, leaving the residual paths to the lower
main concern sub streams. The likely delay is exact by the number of heart nodes of the paths (the less is the
number of intermediate nodes; lower end-to-end delay is achieved). The writer of cite an attractive statement
scenario for the proposed solution. In fire monitor, visual in order is more applicable for the request and should
be deliver with smallest program stoppage. The audio stream could be transmitting over the lasting paths,
complementing the conservative visual data. In live out, as some obtainable paths may have higher broadcast
delays than the time constraint of the application, they are not careful by usual single account coding
applications (acoustic and picture mutually). On the other hand, even accessible paths with high delays might be
used by the request for transmission of the lower-priority sub flow in, maximize the possible statement
throughput. The concept of context-aware multimedia-based cross-layer optimization presented in is further
investigated in. In both works, the original data (72 kbps) is split into an image stream (48 kbps) and an audio
stream (24 kbps). A generic treatment of multistream multipath transmission is given in. Li et al. exploit
multiple images coding for multipath-selection. However, while TPGF was in a job in to create manifold nodedisjoint paths, the work in extended the direct circulation protocol to learn multiple node-disjoint paths. This is
in fact the main contribution of, since MDC is working just to assess the presentation of the predictable
multipath routing procedure. The work obtainable in also combines MDC with multipath transport. However, in
30
A. Sivagami et al,2014
Advances in Natural and Applied Sciences, 8(17) Special 2014, Pages: 23-35
a different way from, the proposed scheme constantly monitors the path to decide the number of sub streams
that have to be created. Among the three video transport method proposed in, only one is fitting for WMSN,
since it does not required advice messages to be sent for each frame by the receiver end, which could add to the
power operation and the end-to-end delay (Shu, 2010). However, as the energy use for any of the future schemes
was not evaluated in this work, their using in visual sensor network may be not feasible, although their help
have inclined other factory.
Coding techniques:
Predictive technique:
As described earlier, the knowledge of the available energy reserves in each part of the network is important
information for sensor networks. The more natural way of thinking about the energy map construction is one in
which periodically each node sends to the monitoring node its available energy. We call this the naïve approach.
As the sensor networks have lots of nodes with limited resources, the amount of energy spent in the naive
approach will be prohibitive. For that reason, better energy efficient techniques have to be designed to gather the
information about the available energy at each part of a sensor network. The possibilities of constructing the
energy map using prediction-based approaches. Basically, each node sends to the monitoring node the
parameters of the model that describes its energy drop and the monitoring node uses this information to update
locally the information about the available energy at each node. The motivation that guided us to this work is
that if a node is able to predict the amount of energy it will spend, it can send this information to the monitoring
node and no more energy information will be sent during the period that the model can describe satisfactorily
the energy dissipation (H. Kimand, 2009). Then, if a node can efficiently predict the amount of energy it will
dissipate in the future time, we can save energy in the process of constructing the energy map of a sensor
network. In order to predict the dissipated energy, we studied two models in which the energy level is represented by a time series and the ARIMA (Autoregressive Integrated Moving Average) model is used to make
the predictions.
i) Probabilistic Mode: In this section, we claim that each antenna node can be modeled by a Markov chain. In
this case, the node modes of operation are represented by the state of a Markov chain and the chance variables
represent the likelihood of staying at each state in a certain time. Then, if each sensor node has M mode of
operation, each node will be modeled by a Markov chain with M states. Using this model, at each node, we have
a run of random variables X0; X1; X2; that represents its states during the time. Then, if Xn = i, we say that the
sensor node is in mode of operation i at time-step3 n. In adding, at each time the node is in state i, there is some
fixed probability, Pij , that the next state will be j. We can also do the n-step transition probability, P(n) ij , that a
node presently in state i will be in state j after n additional transitions With the knowledge of the probabilities
P(n) ij for all nodes and the value of X0 (initial state of each node), it is possible to estimate some information
about the network that can be useful in many tasks. In this work, we will use this probability to predict the
energy drop of a sensor node. The first step to make this calculation is to calculate for how many time-steps a
node will be in a state s in the next T time-steps. If the node is in state i (X0 = i), the number of time-steps a
node will stay in the state s can be designed by: PT t=1. Also, if Es is the amount of energy sink by a node that
remains one time-step in state s, and the node is at present in state i, then the likely quantity of energy spent in
the next T times. Using the value ET, each node can calculate its energy uprightness rate (¢E) for the next T
time-steps. Each node then sends its available energy and its ¢E to the monitoring node. The monitor node can
maintain a view for the dissolute energy at each node by falling the value ¢E sometimes for the amount of
residual energy of each node. The better the estimation the node can do, the fewer the number of letters
necessary to obtain the energy information and, as a result, the less the amount of energy spent in the
development of getting the energy map (B. Henty, 2001).
ii) Statistical Mode: In this section, we present the statistical model used to predict the energy level in the sensor
nodes. In this representation, we represent the energy drop of a sensor node as a time series. A time series is a
set of observations xt, each one being record at a specific time t .A discrete- time series is one in which the set
T0 of times at which observations are made is a discrete set. Continuous-time time series are obtained when
observations are recorded continuously over some time interval. There are two main goals of time series
analysis: identifying the nature of the fact represent by the run of notes, and forecasting (predicting future values
of the time series variable). In this work, we are interested in using the time series analysis to forecast future
values of the available energy in a sensor node. We motivation use the discrete-time time series in such a way
that each node will verify its energy level in a discrete time interval. We can observe that the time chain which
represent the energy drop of a sensor node has a clear decreasing trend4 (we suppose that there is no
replacement in the battery) and no seasonality. The decreasing trend will also imply in a declining mean and
then the energy level will also have no stationary differences are usually sufficient to obtain a stationary time
series. The number of differencing applied in the original series is representing by the parameter d. The next
step in the construction of the ARIMA model is to recognize the AR terms. An autoregressive model is simply a
linear failure of the current value against one or more previous values of the series. The value of p is called the
31
A. Sivagami et al,2014
Advances in Natural and Applied Sciences, 8(17) Special 2014, Pages: 23-35
order of the AR model. Then, an autoregressive model of order p can be summarized by: Xt =
Á1Xt¡1+Á2Xt¡2+:::+ÁpXt¡p+Zt, where Xt is the time series, Á1; Á2; :::; Áp are the autoregressive model
parameters, and Zt represents normally spread random errors. After important the differencing and the
autoregressive parameters, we have to identify the MA terms. A moving standard model is essentially a linear
decay of the current value of the series against the casual shocks of one or more prior values of the series. The
random shocks at each point are assumed to come from the same distribution, typically a normal distribution,
with constanlocation and scale. The distinction in this model is that these random shocks are propagated to
future values of the time series. A moving regular model of order q is represented by: Xt = Zt +µ1Zt¡1 +µ2Zt¡2
+::: +µqZt¡q, where Xt is the time series, µ1; µ2; :::; µq are the moving average model parameter and the Zt
are random shocks to the series. Then, in order to use the ARIMA model we have to identify the values of p
(order of the autoregressive model), d (number of differencing required achieving stationary), q (order of the
moving average model) and the coefficients of the autoregressive and moving usual models. Thus, a time series
Tt can be represented by an ARIMA (p,d,q) model if, after differencing this series d times, we ¯nd a stationary
time series Xt, such that for every t: Xt = Á1Xt¡1 + ::: + ÁpXt¡p + Zt + µ1Zt¡1 + ::: + µqZt¡q. When using
equation above, we can predict the value of the time series in time t using the previous values and some random
variables that represent the errors in the series. In general, the estimation of these parameters is not a trivial task.
In, the authors describe some techniques to help in the process of parameters identification. In this work, we will
use the ARIMA (Autoregressive Integrated Moving Average) model to predict future values of the time series.
The ARIMA model was proposed by Box and Jenkins and they consist of a systematic methodology for
identifying and estimating models that could incorporate both autoregressive and moving average approaches.
This makes ARIMA models a powerful and general class of models. The included part of the model is due to the
differencing step necessary to make the series stationary. The first step in developing an ARIMA model is to
determine if the series is stationary. When the unique series is not stationary, we need to difference it to achieve
stationary. Given the series Zt, the differenced series is a new series Xt = Zt ¡ Zt¡1. The differenced data contain
one less point than the original one.
Multipath Routing in WMSN:
This sector describe the steering operation of our planned cluster multipath steering protocol for WMSNs,
CMRP, which is bottom on the hierarchical arrangement of frequent trail recognized depending on hop count
and established signal strength (along with measured SNR and BER) as an signal of the link quality and distance
between the nodes. CMRP depends on the local in rank trade among the nodes to set up the routes to the fall and
does not require any organization dimension equipments or place message swap.
a) Route Discovery: The two routine metrics are: hop count (as sign for coldness from the descend and holdup)
and conventional signal strength index RSSI (joint with SNR & BER) as sign for link excellence (obstruction
and noise level) and detachment from the dispatcher. Two thresholds (upper and lower) are used to match up to
with the packet’s RSSI. The group of the values of the two thresholds is very critical in move toward together
the system and between them together. The upper entry is used to determine the 1st level group heads and group
associate nodes (as explained below). The upper edge be hypothetical to be used to in a way that it should not be
very large value (close to the max value) so that you will not find any node receives your letter in this power
level or only a few nodes. In this case, the group size will be very small with many probability of have only
singleton group and the load will be high on few 1stlevel cluster heads for serving many paths passing through
them. Also if the upper threshold is low (below the mid value close to the lower threshold), the come together
size will be very high and cause group heads to load with many group members and experience high
interferences in both inside cluster and at the drop side. The lower threshold is used to set up the relations
between group head. have a quite high value of the lower entrance (close to the mid value) may prevent
connecting the collect heads in dissimilar levels and this leads to have weak scheme connectivity. Also if the
inferior entry is very low (close to the mid value), then the network can have low link quality links between
cluster heads with high option of packet drops. In the initializing phase of CMRP, the base station starts sharing
broken up broadcast messages, called BS-Msg, to the surrounding powerful nodes. The nodes that receive BS
Msgs contrast the RSSI with the upper threshold (Thr- High). If RSSI is better than Thr-High, these nodes
counter to the base station by sending back admission messages informing their joining the base station as their
parent. Then, they start acting as 1st-level cluster heads (1st CH) and broadcast once in a while control letters
called CH-Msg to their adjacent nodes. CH-Msg contains the ID of the CH, number of hops between the CH and
the sink in the present found path, and IDs of the nodes union this path up to the current CH. For each CH-Msg
received by the instant nodes of the 1st CHs, RSSI is measured and compare with the two thresholds, Thr-High
and Thr-Low. If the signal strength of the received CH-Msg is greater than Thr-High, the in receipt of node will
start behaving as a group member (GM) and send back an credit message tell its joining to the identical CH.
Receiving a CH-Msg with RSSI better than Thr-High indicates that the sender (CH) is in near region and the
excellence of the link is good, and thus, this CH can better serve the statement toward the base station. In case a
node receive more than one message from different CHs with RSSI better than Thr-High, the node choose the
32
A. Sivagami et al,2014
Advances in Natural and Applied Sciences, 8(17) Special 2014, Pages: 23-35
cluster head of the highest RSSI value, as shown in pseudo code of CMRP in Algorithm. The influential nodes
that only receive messages with RSSI between Thr-High and Thr-Low will start acting as new cluster heads, in
this case 2nd-level CHs, and respond back to the sender informing their assortment of him as one of their
possible parents toward the base station. New CHs may receive different CH-Msgs from previous-level CHs. In
this case, new CHs consider these mail in order to construct multiple paths on the way to base station and sort
these paths based on certain measure (such as link quality, end to- end delay, bandwidth, or number of hops in
the path). Paths with good environment, like high link quality, short end to-end holdup, enough bandwidth, or
less number of hops, are kept for CD speech that requires positive level of quality of service supplies. Other
paths will be used for other types of data that does not require strict QoS supplies such as scalar data. If the
RSSI is less than Thr-Low, the mobile phone call is considered as lost or unobserved. This process carry on in
the same style to build the scheme until all nodes join the network and end their roles, that is, collect head or
group link, and all likely path are found.
Distributed Coding Structure:
Today the complex surrounds significantly differ from the one the TCP/IP reference model was designed
for. In this heterogeneous setting, TCP/IP shows poor performance, driving innovation towards the naming of
more cooperative cross-layer design solutions. The idea of spread procedure Stacks is built up on cross-layering
combined with “agent-based networking” and proxy-based design. The Distributed Protocol Stacks architecture
extends traditional layered (ISO/OSI or TCP/IP) protocol stack by allowing abstractions of “atomic” functions
from a specific protocol layer and by given that means of detaching the inattentive efficient blocks from the
protocol stack in order to move them within the network (“in-network”). The act of relayed transmission can be
further optimized if joint signal design and coding are performed at the source and relays.
a) Distributed Space Time Coding: Let us consider a wireless pass on network in which the foundation and
convey kindly communicate with a common destination. This helpful broadcast among the basis and relays
forms a virtual antenna array. Therefore, conventional space-time coding schemes can be applied to relay
networks for attain the supportive variety and coding gain. Two types of distributed space -time coding (D-STC)
schemes have been urban, including distributed space-time block codes (DSTBCs) and disseminated space time
lattice codes (DSTTCs). The alteration is unique for each relay and is represented by a signature vector. It has
been shown that by correct code design, DSTBC can achieve a range order equal to the number of active relay
nodes Several DSTBC schemes have been future. A simple DSTBC scheme was projected by Laneman based
on orthogonal STBCs (E. Hossain, 2004). To solve this problem, several solutions have been projected. In
future a DSTBC scheme, which selects a subset of nodes for show, and each active relay nodes.
Table 1: Comparison of Existing approaches for cross layer architecture in WMSN.
Protocol
Network modality
Operational Layer
Costa and L. A.
Standard
Routing
Guedes,(2011).
Kompella, S. Mao, Y. T. Hou,
Cross- Layer
Routing/MAC
and H. D. Sherali,(2007).
Srivastava and M.
Cross- Layer
Routing
Motani,(2005).
Shakkottai, T. S. Rappaport
Cross- Layer
Applications/Routing
and P. C. Karlsson(2003),
Pollin, B. Bougard, and G.
Cross- Layer
Routing/MAC
Lenoir,(2003).
Sharif and B. Hassibi,(2005)
Cross- Layer
Transport Routing
Shiang and M. Var der
Cross- Layer
Application/Routing/MAC
Schaar,(2009)
Performance Metrics
Transmission Count/delay
Bandwidth/delay
Hop Count/delay
Bandwidth/delay/path energy
End-to-to energy reservation
Distance/delay/data type
Bandwidth/delay/distortion
Conclusion:
Due to the great success of wireless multimedia application and wireless mobile connections, there has been
a dramatic demand for wireless data access. Within a layered architecture, it is possible to yield significant gains
if the system optimizes the performance by making use of the interaction across different protocol layers. Thus,
in this thesis, focus is made on employing cross-layer technique for preparation and resource part in wireless
networks. The main objective is to improve the system concert by incorporating the information from the
physical layer and MAC layer into the design of resource management. The conclusion arrived from the
research work carried out and the future scope of the work is discussed in this chapter.
Contributions of the research:
The study of cross-layer and resource share performed on wireless networks brought out the following
outcome and conclusions:
(i) In order to assurance QoS happiness, an Adaptive Cross-layer Packet Scheduling (ACPS) algorithm is
proposed for wireless networks under IEEE 802.11b standard. This algorithm adapts to change in variables like
33
A. Sivagami et al,2014
Advances in Natural and Applied Sciences, 8(17) Special 2014, Pages: 23-35
package delay on the link layer and channel bit rates on the physical layer across two system. This accurately
characterizes the influence of physical layer infrastructure on diverse QoS performance at the higher protocol
layers. The cost function employed in the ACPS algorithm is a weighted estimate of the average normalized
packet delay of packet in the queue. This algorithm attempts to minimize the average packet delay at every
chance of development. Also by allow high bit rate packets with more urgency through the channel for spread at
good SNR conditions, this algorithm is achieving more throughput than the conventional WFQ algorithm. By
comparing ACPS algorithm with WFQ algorithm, it is shown that ACPS outperforms WFQ when 64 bytes sized
packets are used with any priority class file. When the packet size is 128 bytes, the average throughput feat
improvement in ACPS algorithm is very minimal. For 192 bytes packet sizes, WFQ algorithm is presentation
better average user throughput performance than that of ACPS algorithm. It is because of the larger delay
experienced, completion of maximum allowable lifetime in the network and more loss of packets in ACPS
algorithm. For example, copy results show that the average packet delay time taken by P3 class files with 64
bytes packets using ACPS algorithm (950 s) is nearly 21 times lower than that of the average packet delay time
taken by 64 bytes packets using WFQ algorithm, (20 ms) equally delay of P1 file with 64 bytes packets is 21 s,
which is nearly 950 times lower than that of the delay experienced by WFQ algorithm, i.e., 20 ms. thus the
outperformance of the projected ACPS algorithm for all three priority classes than WFQ algorithm is verified.
(ii) The effectual capacity based cross-layer scheduling algorithm is developed to support the real-time
multimedia QoS by allocating channel capital adaptively for downlink heterogeneous mobile wireless network
under IEEE 802.16 standard. According to the channel fading statistics, the diverse QoS requirements and
traffic characteristics, the time slots are allocated as system resource to the users(George Xylomenos,2001). The
problem of physical layer impact on the statistical QoS provisioning performance and the influence of adaptive
channel state information feedback delay on the proposed algorithm are studied. From the numerical results, it is
understood that the effective capacity based cross-layer scheduling algorithm is providing significant (for e.g., 6
audio time slots and 5 video time slots) reduction in the number of time period for the similar type of services
over conformist constant power approach. As the SNR increases, normally the number of time slots taken or
assigned for transmission decreases and vice versa. Thus the proposed effective capacity based scheduling
algorithm reveals better significance in terms of minimum number of time slots allocated to the audio and video
services over conventional constant power come close to.
Scope for future research:
In addition to the spectral competence, the other parameter justice and QoS are crucial for resource share in
wireless networks. It is often difficult to achieve the optimality for spectral efficiency, fairness and QoS
simultaneously (A. J. Goldsmith,1998). For instance, scheduling schemes aiming to make the most of the total
throughput are unfair to those users far away from a base station or with bad channel conditions. On the other
hand, the absolute fairness may lead to low bandwidth efficiency. Then, an effective trade-off among efficiency,
fairness and QoS are desired in wireless resource allocation. It would be of great interest to extend our research
towards the development of such swapping scheduling algorithms. Also, this thesis mainly considers multipleaccess channels where the full coordination at the base-station is feasible. Of great interest is to study crosslayer resource allocation in interference channels or relay channels without full organization (T. S. Rappaport,
2002). Interference channels can be encountered in many applications such as multi-cell wireless networks and
DSL networks. Also, relay channels can be found in a multi-hop ad hoc wireless network where multiple hops
are required for efficient communication between mobile nodes far apart as referred in Goldsmith et al (2002).
With no or partial supervision, it becomes more testing to satisfy each user's different QoS demand and to
enlarge efficient power / rate optimization algorithms.
REFERENCES
Akyildiz, T. Melodia and K.R. Chowdhury, 2007. A survey on wireless multimedia sensor networks,
Computer Networks, 51(4): 4, 921–960.
Ali, M. Hayajneh and H. Hassanein, 2008. Cross layer scheduling algorithm for IEEE 802.16 broadband
wireless networks: in IEEE International Conference on Communications, pp: 3858–3862.
Almalkawi, M. Guerrero Zapata, J.N. Al-Karaki and J. Morillo-Pozo, 2010. Wireless multimedia sensor
Networks: current trends and future directions. Sensors, 10(7): 6662– 6717.
Begen, Y. Altunbasak, O. Ergun and M.H. Ammar, 2005. Multi-path selection for multiple description
video streaming over overlay networks,Signal Processing: Image Communication, 20(1): 39–60.
Boulis, M. Srivastava, 2003. Node-level energy management for sensor networks in the presence of
multiple applications, in: Proc. of IEEE Intl. Conf. on Pervasive Computing and Communications (PerCom),
Dallas Forth Worth, TX, USA, pp: 41-49.
34
A. Sivagami et al,2014
Advances in Natural and Applied Sciences, 8(17) Special 2014, Pages: 23-35
Cayman Khalil, Mathieu Crussiere and Jean-Francis Helard, 2010. Cross Layer Resource allocation scheme
under heterogeneous constraints for next generation high rate WPAN. International Journal of Computer
Networks and Communications (IJCNC), 2(3): 152-168.
Chen, S.H. Low and J.C. Doyle, 2005. Joint Congestion Control and Media Access Control Design for Ad
Hoc Wireless Networks, Proc. INFOCOM.
Chen, V. Leung, S. Mao and M. Li, 2008. Cross-layer and path priority scheduling based real-time video
communications over wireless sensor networks, in Proceedings of the IEEE Vehicular Technology Conference
(VTC ’08): 2873–2877, Marina Bay, Singapore.
Costa and L.A. Guedes, 2011. A survey on multimedia-based cross-layer optimization in visual sensor
networks,Sensors, 11(5): 5439–5468.
Daeyoung Park, H. Seo, Hojoong Kwon and Byeong Gi Lee, 2005. Wireless packet scheduling based on
the cumulative distribution function of user transmission rates, IEEE Transactions on Communications, 53(11):
1919-1929.
Ehsan and B. Hamdaoui, 2011. A survey on energy-efficient routing techniques with QoS assurances for
wireless multimedia sensor networks, IEEE Communications Surveys and Tutorials, 99: 1–14.
George Xylomenos and George C. Polyzos, 2001. Quality of Service Support over Multi-Service Wireless
Internet Links. Computer Networks, 37(5): 601–615.
Goeckel, 1999. Adaptive coding for time-varying channels using outdated fading estimates, IEEE Trans
Commun., 47: 844–855.
Goldsmith and S.G. Chua, 1998. Adaptive coded modulation for fading channels, IEEE Trans. Commun.,
46(5): 595–602.
He, 2003. SPEED: A Stateless Protocol for Real-Time Communication in Sensor Networks, In Proc. of Int.
Conference on DCS.IEEE Standard 802.16 Working Group(2012, IEEE Standard for Local and Metropolitan
Area Networks Part 16: Air Interface for Fixed Broadband Wireless Access Systems.
Henty and T.S. Rappaport, 2001. Throughput measurements and empirical prediction models for IEEE
802.11b wireless LAN (WLAN) installations,Tech. Rep. MPRG 01-08, Dept. of Electrical and Computer
Engineering, Virginia Tech, Blacksburg, VA.
Hossain and V.K. Bhargava, 2004. Link-level traffic scheduling for providing predictive QoS in wireless
multimedia networks, IEEE Trans. Multimedia, 6(1): 199–217.
Hou, P.H. Ho and X. Shen, 2006. A novel differentiated retransmission scheme for MPEG video streaming
over wireless links,Int. J. Wireless and Mobile Computing, 1(3/4): 260-26.
Kim and M.M. Krunz, 2000. Delay analysis of selective repeat ARQ for Markov chain source over a
wireless channel, IEEE Trans. Veh. Technol, 49(5): 1968-1981.
Kimand, S.G. Min, 2009. Priority-based QoS MAC protocol for wireless sensor networks, in Proc. IEEE
IPDPS’09, pp: 1–8.
Kompella, S. Mao, Y.T. Hou and H.D. Sherali, 2007. Cross layer optimized multipath routing for video
communications in wireless networks, IEEE Journal on Selected Areas in Communications, 25(4): 831–840.
Kumwilaisak, Y. Thomas Hou, Q. Zhang, W. Zhu, C.C. Jay Kuo, Ya-Qin Zhang, 2003. A cross-layer
quality of service mapping architecture for video delivery in wireless networks, IEEE Journal on Selected Areas
in Communications, 21(10): 1685-1698.
Kunniyur and R. Srikant, 2000. End-to-end Congestion Control: Utility Functions, Random Losses and
ECN Marks, Proc. IEEE INFOCOM, 3: 1323–32.
Lecuire, V., C. Duran-Faundez, N. Krommenacker, 2004. Energy-efficient image transmission in sensor
networks. Int. J. Sens. Netw. 2008, 4, 37-47. 26 F. Aune, Cross-Layer Design Tutorial, Norwegian University of
Science and Technology, Dept. of Electronics and Telecommunications, Trondheim, Norway.
Licrandro, F., A. Lombardo, G. Schembra, 2008. Multipath routing and rate-controlled video encoding in
wireless multimedia sensor networks. Multimed. Syst., 14: 155-165.
Lin and N.B. Shroff, 2005. The Impact of Imperfect Scheduling on Cross Layer Rate Control in Wireless
Networks, Proc. INFOCOM,
Liu, S. Zhou and G.B. Giannakis, 2004. Cross-layer combining of adaptive modulation and coding with
truncated ARQ over wireless links,IEEE Trans. Wireless Commun., 3(5): 1746–1755.
Oldewurtel, J. Ansari and P. Mahonen, 2008. Cross-layer design for distributed source coding in wireless
sensor networks,in Proc. SENSORCOM, pp: 435–443.
Othman and B. Yahya, 2010. Energy efficient and QoS based routing protocol for wireless sensor networks,
J. Parallel Distrib. Comput., 70(8): 849–857.
Physical layer aspects of ultra high speed downlink packet access (release 4),3GPP TR 25.848 V4.00,2001.
Physical Layer Standard for CDMA2000 Spread Spectrum Systems(1999). 3GPP2 C.S0002–0 Ver. 1.0.
Pollin, B. Bougard and G. Lenoir, 2003. Cross-Layer Exploration of Link Adaptation in Wireless LANs
with TCP Traffic,Proc. IEEE Benelux Chapter on Communications and Vehicular Technology.
35
A. Sivagami et al,2014
Advances in Natural and Applied Sciences, 8(17) Special 2014, Pages: 23-35
Raisinghani and S. Iyer, 2006. Cross Layer Feedback Architecture for Mobile Device Protocol Stacks,
IEEE Communications Magazine, 44(1): 85–92.
Rappaport, A. Annamalai, R.M. Buehrer and W.H. Tranter, 2001. Wireless communications: Past events
and a future perspective, IEEE Communications Magazine, 40(148)–1MAY[46] Physical Layer Aspects of
UTRA High Speed Downlink Packet Access (Release 4), 3GPP TR 25.848 V4.0.0, .
Rappaport, Wireless Communications, 2002. Principles and Practice. Upper Saddle River, NJ: Prentice
Hall.
Shakkottai, T.S. Rappaport and P.C.K arlsson, 2003. Cross-layer design for wireless networks, IEEE
Communications Magazine, 41(10), October 2003, 74-80.S.
Sharif and B. Hassibi, 2005. A delay analysis for opportunistic transmission in fading broadcast channels,
Proc. IEEE INFOCOM,.
Shiang and M. Var der Schaar, 2009. Distributed resource management in multi-hop cognitive radio
networks for delay sensitive transmission,IEEE Trans. Veh. Technol., 58(2): 941–953.
Shu, L., Y. Zhang, Z. Yu, L.T. Yang, M. Hauswirth, N. Xiong, 2010. Context-aware cross-layer optimized
video streaming in wireless multimedia sensor networks. J. Supercomput, 54: 94 121.
Shu, M. Hauswirth, L. Wang, Y. Zhang, and J.H. Park, 2009. Cross-layer optimized data gathering in
wireless multimedia sensor networks, in Proc. CSE’09: 961–966.
Siteplanner, Wireless Valley Communications inc..http://www.wirelessvalley.com. Also see US Patents
6499006, 6493679, 6442507 and 6317599
Srivastava and M. Motani, 2005. Cross-layer design: a survey and the road ahead, IEEE Communications
Magazine, 43(12): 112–119.
Tarabanis, K., P. Allen, R. Tsai, 1995. A Survey of sensor planning in computer vision. IEEE Trans. Rob.
Autom., 11: 86-104.
Toumpis and A.J. Goldsmith, Regions, 2003. capacity for Wireless Ad Hoc Networks, IEEE Transactions
on Wireless Communications, 2(4): 746-748.
Triantafyllopoulou, N. Passas and A. Kaloxylos, 2007. A cross-layer optimization mechanism for
multimedia traffic over IEEE 802.16 networks, International Journal of Network Management,17( 5),
September, pp: 347–361.
Villal on, P. Cuenca, L. Orozco-Barbosa, Y. Seok and T. Turletti 2007. Cross-layer architecture for
adaptive video multicast streaming over multirate wireless LANs,IEEE Journal on Selected Areas in
Communications, 25(4): 699–711.
Wang and M.A. Abu-Rgheff, 2003. Cross-layer signaling for next-generation wireless systems, IEEE
Wireless Communications and Networking (WCNC), pp: 1084–1089.
Xiong, 2004. Distributed Source Coding for Sensor Networks, IEEE Signal Processing Magazine,. 80-94, .
Xu, V. Stankovi´c and Z. Xiong, 2007. Distributed joint source channel coding of video using raptor codes,
IEEE Journal on Selected Areas in Communications, 25(4): 851–861.
Zhang, S.L. Regunathan and K. Rose, 2000. Video coding with optimal inter/intra-mode switching for
packet loss resilience, IEEE journal on selected areas in communication, 18(6): 966-976.
Zhu,W. Zeng and C. Li, 2007. Cross-layer design of sources rate control and congestion control for wireless
video streaming, Advances in Multimedia, Hindawi publishing corporation,1 article ID 6850, 3-15.